基于电商评论的用户需求挖掘研究：以助听器为例
Research on User Demand Mining Based on E-Commerce Reviews: Taking Hearing Aids as an Example
摘要: 研究基于京东助听器用户在线评论文本数据，对比机器学习算法与双向长短时记忆神经网络(BiLSTM)的模型性能，发现BiLSTM模型在用户情感分类上性能最佳，准确率达90.25%，F1 Score为72.3%。研究在应用BiLSTM分析评论情感极性后，借助主题关键词提取技术(LDA)，识别当前助听器用户的需求，为助听器的持续改进和创新提供依据。
Abstract: Based on online review text data from JD hearing aid users, this study compared the performance of machine learning algorithms and a bidirectional long short-term memory neural network (BiLSTM) model. It was found that the BiLSTM model performed best in user sentiment classification, achieving an accuracy rate of 90.25% and an F1 Score of 72.3%. After applying BiLSTM to analyze review sentiment polarity, the study utilized topic keyword extraction technology (LDA) to identify the needs of current hearing aid users, providing a basis for continuous improvement and innovation of hearing aids.
文章引用：李欣雨, 罗鄂湘, 贾泽如. 基于电商评论的用户需求挖掘研究：以助听器为例[J]. 电子商务评论, 2026, 15(2): 338-347. https://doi.org/10.12677/ecl.2026.152164

参考文献

[1] 李高翡, 张洋, 杨新凯, 等. 基于集成学习的文本情感分析研究[J]. 计算机应用研究, 2020, 37(S1): 50-51.
[2] 孙玲玲, 胡彦蓉, 刘洪久. 基于LSTM-LDA算法和IPA分析的在线品牌社群用户关注热点研究[J]. 情报杂志, 2021, 40(9): 178-186.
[3] 李俭兵, 刘栗材. 基于改进型神经网络的影评文本情感分析算法[J]. 计算机工程与科学, 2019, 41(12): 2261-2269.
[4] 季陈, 叶慧雯, 王子民, 等. 基于BiLSTM-CapsNet混合模型的社交文本情感分析[J]. 南京理工大学学报, 2022, 46(2): 205-210.
[5] 张国方, 寇姣姣, 陈令华. 网络评论文本驱动的汽车设计规划方法[J]. 机械设计, 2021, 38(2): 139-144.
[6] 周玲玲, 周晶, 韩朝, 等. 助听器对伴听力损失的慢性主观性耳鸣的临床效果分析[J]. 中国耳鼻咽喉头颈外科, 2025, 32(2): 90-93.
[7] 叶欣, 朱大伟, 高嘉敏, 等. 中国老年人听力对抑郁的影响研究[J]. 人口与发展, 2019, 25(5): 66-74.
[8] 常东旭, 徐春燕, 孟坤光, 等. 老年性听力损失患者助听器配戴效果研究进展[J]. 中国听力语言康复科学杂志, 2023, 21(6): 626-629.
[9] WHO (2021) World Report on Hearing.
https://www.who.int/publications/i/item/9789240020481
[10] 宫蕊, 卜行宽. 全国部分省听力障碍流行现状和康复需求及服务利用情况[M]//中国听力健康报告(2021). 北京: 社会科学文献出版社, 2021: 72-83.
[11] Hochreiter, S. and Schmidhuber, J. (1997) Long Short-Term Memory. Neural Computation, 9, 1735-1780. [Google Scholar] [CrossRef] [PubMed]
[12] da Silva, D.G. and Meneses, A.A.D.M. (2023) Comparing Long Short-Term Memory (LSTM) and Bidirectional LSTM Deep Neural Networks for Power Consumption Prediction. Energy Reports, 10, 3315-3334. [Google Scholar] [CrossRef] 
[13] Blei, D.M., Ng, A.Y. and Jordan, M.I. (2003) Latent Dirichlet Allocation. Journal of Machine Learning Research, 3, 993-1022.
[14] Ma, J., Wang, L., Zhang, Y., Yuan, W. and Guo, W. (2023) An Integrated Latent Dirichlet Allocation and Word2vec Method for Generating the Topic Evolution of Mental Models from Global to Local. Expert Systems with Applications, 212, Article 118695. [Google Scholar] [CrossRef] 
[15] Zimmermann, J., Champagne, L.E., Dickens, J.M. and Hazen, B.T. (2024) Approaches to Improve Preprocessing for Latent Dirichlet Allocation Topic Modeling. Decision Support Systems, 185, Article 114310. [Google Scholar] [CrossRef] 
[16] Guntzburger, Y. and Hadengue, M. (2025) Providing a Comprehensive Thematic Review of the Science-Policy Interface (SPI): A Probabilistic Topic Modeling Approach. Environmental Science & Policy, 163, Article 103966. [Google Scholar] [CrossRef] 
[17] Yu, D. and Xiang, B. (2023) Discovering Topics and Trends in the Field of Artificial Intelligence: Using LDA Topic Modeling. Expert Systems with Applications, 225, Article 120114. [Google Scholar] [CrossRef] 
[18] 刘洪波, 梁一帆, 刘虹. 助听设备产品外观设计趋势研究[J]. 包装工程, 2022, 43(S1): 10-15.
[19] 杨超翔, 原炳坤, 叶俊男. 基于模糊QFD的红外理疗仪适老化设计方法研究[J]. 机械设计, 2022, 39(3): 147-153.

为你推荐